The efficacy of TNF antagonists and B cell depletion therapy (BCDT) in rheumatoid arthritis (RA) have fundamentally altered pathophysiologic paradigms and raised new quesfions about disease mechanisms. The observation that both TNF and B cells are of central importance provides a strong impetus to improve our understanding ofthe relationship between TNF over-production and infiammation pathways mediated by B cells, particulady in regards to disease flares and inadequate treatment response. The TNF-transgenic (TNF-tg) mouse strain is the only chronically progressive model for RA with a clinically proven etiology, and thus invaluable to address the cellular and immunological events in RA. We have developed a series of novel small animal imaging approaches, including contrast enhanced (CE) MRI, in vivo micro-CT and nearinfrared indocyanin green (NIR-ICG) analysis of lymphatic flow, to study longitudinally the events associated with progression of joint arthritis. The results from these studies, combined with histology and flow cytometry, showed dramatic changes unfolding in a predictable temporal sequence in the draining popliteal lymph-node (PLN) before and duhng the onset of knee disease. Pnor to knee synovitis, the draining PLNs "expand", as evidenced by increased volume and CE values (i.e. lower density), and histology of the node shows markedly dilated, enlarged sinusoids. This nodal expansion is also characterized by increased PLN B cell numbers, primanly due to accumulation of a B cell subset with a unique surface phenotype ("B-in" cells). This phase is followed by a rapid decrease in node size and CE, by invasion and "plugging" ofthe lymphatic sinuses by B cells, and reduced lymphatic flow through the node. The onset of knee synovitis and focal erosions follows PLN collapse. Both BCDT and anti-CXCL13 treatment result in TNF-tg disease ameliorafion and recovery of PLN size, CE and histology, pointing to a pathogenetic role of both B cells and chemotaxis. Thus, we hypothesize that evolution of knee arthritis and PLN function in TNF-tg mice proceed from an early stage during which pathogenetic B cells are recruited to and accumulate in the draining nodes, which increase in size and CE. At a later stage PLN-intrinsic or external (e.g. from the joint) signals induce migration ofthe expanded B cell population to the sinusoidal spaces, resulting in the obstruction of lymph flow and changes in the local cytokine balance. This leads to accumulation of pro-inflammatory mediators in the knee joint, triggering the "flare". Here we propose to test this model by: 1) identifying the mechanisms that lead to B cell accumulation in the PLNs and their subsequent migration to lymphatic spaces;2) demonstrating the pathogenetic role of adoptively transfered PLN B cells, and testing the hypothesis that BCDT efficacy is associated with clearance of sinusoid-invading B cells;3) testing whether TNF-tg B cells can exert pro-inflammatory effector functions;and d) investigating the applicability of these finding to RA patients in a pilot clinical tnal in which PLN structure and lymphatic function wil be assessed before and after BCDT, and correlated with clinical svmptoms.